Differences in positive and negative affect across social achievement goal clusters among Japanese college students

Positive affect (PA) and negative affect (NA) are important contributors to college students' successful campus lives; however, the extent to which social achievement goals are associated with these affective states remains underexplored. These goals encompass three orientations toward social competence, namely social development, social demonstration-approach, and social demonstration-avoid goals. The first purpose of this study was to identify typical combinations of the three types of social achievement goals among Japanese college students. The second was to examine between-subgroup differences in PA and NA. An online survey was conducted in June 2024 with 330 Japanese college students (mean age = 20.13 years; 50% female). Participants completed self-report measures assessing social achievement goals as well as daily PA and NA. Cluster analysis using Ward's method was performed. Four clusters emerged: (1) a low cluster, characterized by low levels of all three goal types; (2) an average cluster, characterized by average levels; (3) a high cluster, characterized by high levels; and (4) a low social demonstration-approach goal cluster. The low cluster scored significantly lower on PA than the other clusters and significantly lower on NA than the high and average clusters. Moreover, the low social demonstration-approach goal cluster scored lower on NA than the high cluster. These findings enhance understanding of the relationship between social achievement goal profiles and emotional adjustment among college students.

Both positive affect (PA) and negative affect (NA) are important factors in college students' successful campus lives. PA refers to the feelings of enthusiasm, activation, and alertness, while NA refers to general feelings of subjective distress and unpleasant engagement (Watson et al., 1988). PA and NA have been suggested to be mutually related to some extent but are also independent of each other (Watson et al., 1988). Higher levels of PA (Pressman et al., 2019) and lower levels of NA (Levine et al., 2021) have been consistently associated with improved physical health. Furthermore, they have been reported to be associated with academic performance, whereby higher levels of PA and lower levels of NA correlate with enhanced academic performance (Pekrun et al., 2009). Therefore, it is important to increase our knowledge of the factors that correlate with PA and NA among college students.

Among the various factors, social achievement goals represent one of the factors that a relatively small number of studies have examined in relation to PA and NA. Social achievement goals represent orientations toward social competence, which refers to both social skills and abilities that contribute to acceptance by peers and the formation of friendships (Ryan and Shim, 2006). Ryan and Shim (2006) proposed the social development, social demonstration-approach, and social demonstration-avoid goals. The social development goals focus on the development of social competence, which includes improved social skills, deepened quality of relationships, and a developed social life. The social demonstration-approach goals focus on demonstrating social competence and gaining positive evaluation from others. The social demonstration-avoid goals focus on avoiding demonstrations that reveal one's social competence is insufficient. Several studies (Kuroda and Sakurai, 2011; Shim et al., 2013; Zhao, 2022) have reported the association between social achievement goals and college students' affect. Shim et al. (2013) examined the relationship between social achievement goals, PA, and NA among college students. The results showed that social development goals were associated with higher PA (specifically love and joy); social demonstration-avoid goals with lower levels of joy and higher NA (fear, shame, and sadness); and social demonstration-approach goals with higher levels of joy. However, these studies are limited in their inability to consider the associations between combinations of the three types of social achievement goals and affect. These studies have examined each type of social achievement goal separately. Such an analytical strategy is called a variable-centered approach and cannot consider the effects of the other two types while examining one of the three types (Wang et al., 2021).

Some studies have applied a person-centered approach to address this limitation. Such studies have been conducted from a multiple-goal perspective. This perspective assumes that individuals endorse multiple goals simultaneously (Pastor et al., 2007). Some studies (Lee, 2018; Michou et al., 2016) have classified the participants into two or more subgroups characterized by different combinations of three social achievement goals. These studies consistently found two subgroups characterized by relatively low and high levels of all three goals. In addition to these subgroups, Michou et al. (2016) found a subgroup characterized by relatively high scores of social demonstration-approach goals. Furthermore, Lee (2018) found two subgroups, one characterized by relatively high scores of social development and social demonstration-approach goals and the other by relatively high scores of social demonstration-approach and social demonstration-avoid goals. Lee (2018) examined between-subgroup differences in social anxiety. Participants in the subgroup characterized by relatively high levels of both social demonstration-approach and avoid goals scored significantly higher than those in the other three subgroups. Correlational analyses indicated that, among the three types of social achievement goals, only social demonstration-avoid goals were related to general distress and avoidance (r = 0.25). Interestingly, the level of social demonstration-avoid goals was significantly lower in the subgroup characterized by relatively high levels of both social demonstration-approach and social demonstration-avoid goals than in the subgroup with relatively high levels of all three types of social achievement goals.

The person-centered approach to associations between social achievement goals and affect has the following strengths. First, it can clarify subgroups of college students, each of which has a certain combination of the three types of social achievement goals. Second, it can reveal the combinations among most adaptive college students (Pastor et al., 2007), for example, in terms of affective states. Despite this potential, no study has simultaneously examined PA and NA using a person-centered approach among Japanese college students. Additionally, relatively few have examined these subgroups. Michou et al. (2016) focused on 12- and 15-year-old Turkish students, whereas Lee (2018) focused on Korean high school students. Furthermore, regarding affective outcomes, little is known about the differences of levels of PA and NA across such subgroups. Lee (2018) focused on social anxiety, a form of NA with cognitive and behavioral aspects, but did not examine PA. Since PA and NA are independent constructs (Watson et al., 1988), examining them simultaneously will provide more comprehensive understanding of the relationship between social achievement goals and affect.

The purpose of this study was twofold. The first was to identify the combination of the three types of social achievement goals that Japanese college students typically have. It seems possible to hypothesize that two subgroups may emerge, as follows. Consistent with the studies conducted by Michou et al. (2016) and Lee (2018), it was hypothesized that two subgroups characterized by relatively low and high levels of all three goals would emerge (Hypothesis 1). The second objective was to examine the between-subgroup differences in PA and NA. As there is little information available comparing between-subgroup differences in depression predisposition, it was not possible to establish a hypothesis that included differences between specific pairs of subgroups. Instead, we hypothesized that there would be a significant difference in PA and NA across the subgroups (Hypothesis 2). Because the theoretical mechanisms linking distinct combinations of social achievement goals to PA and NA remain underspecified, the present study is exploratory in nature. This study provides an initial empirical examination of typical goal profiles and their affective correlates. The findings are anticipated to guide the formulation of more precise, theory-driven hypotheses in future research.

Participants and procedures

A web-based survey was conducted in late June 2024 using an online research panel administered by a Japanese survey company. The study design and survey contents were developed by the authors, while the survey company was responsible for participant recruitment and data collection. As part of the screening process, registered panel members were asked whether they were currently enrolled as college students. Individuals who met this criterion were invited to complete the questionnaires described in detail subsequently. Valid responses were obtained from 330 college students (mean age = 20.13 years, SD = 1.57; 165 males and 165 females). The participants included 25.5% freshmen (n = 84), 20.0% sophomores (n = 66), 22.1% juniors (n = 73), 29.7% seniors (n = 98), 0.9% fifth graders (n = 3), and 1.8% sixth graders (n = 6). Regarding participants' regions of residence, the largest proportion resided in the Kanto region (38.5%, n = 127), followed by Kinki region (21.8%, n = 72), Chubu region (13.6%, n = 45), and Kyushu region (10.0%, n = 33). Smaller proportions of participants resided in the Chugoku region (6.1%, n = 20), Tohoku region (5.2%, n = 17), Hokkaido (3.3%, n = 11), and Shikoku region (1.5%, n = 5).

Measures

Social achievement goals

This study used Shinohara's (2019) validated Japanese version of the Social Achievement Goals Questionnaire (Ryan and Shim, 2006). This scale comprises 12 items and three subscales, each with four items. The three subscales are social development, social demonstration-approach, and social demonstration-avoid goals. Each participant was instructed to rate the items on a five-point scale ranging from 1 (not at all true) to 5 (very true). Those who scored higher had more social achievement goals than those who did not. Previous research demonstrated adequate reliability and validity for the Japanese version of this scale (Shinohara, 2019). All subscales exhibited acceptable internal consistency in this study (social development goals: Cronbach's α = 0.74; social demonstration-approach goals: Cronbach's α = 0.83; social demonstration-avoid goals: Cronbach's α = 0.71).

Positive affect and negative affect

PA and NA were assessed using the Japanese version of the Positive and Negative Affect Schedule (PANAS; Watson et al., 1988), which was translated and validated by Sato and Yasuda (2001). The PA and NA subscales each comprise eight items. Participants rated the extent to which they experienced each affective state in their daily lives life on a six-point Likert scale ranging from 1 (not at all true) to 6 (very true), with higher scores indicating greater levels of positive or negative affect. Previous research has supported the reliability and validity of the Japanese version of the PANAS (Sato and Yasuda, 2001). In the present study, both subscales showed good internal consistency (PA: Cronbach's α = 0.91; NA: Cronbach's α = 0.92).

Ethical considerations

At the beginning of this survey, the purpose of the study, management of responses, and handling of responses were explained on-screen. Only those who provided consent were asked to complete the survey. This study was approved by the Research Ethics Committees of Hijiyama University and Hijiyama University Junior College (Application No. 2403; approved on June 21, 2024).

Statistical analysis

SPSS for Windows (version 28) was used for statistical analysis. The significance level was set at p < 0.05. Cluster analysis (Ward's method) was conducted to divide the participants based on the standardized scores of the three social achievement goals. Prior to conducting the cluster analysis, the distributions of the clustering variables were examined, and no extreme outliers were detected. Two one-way analyses of covariance were conducted with PA and NA as dependent variables and sex as the covariate. The independent variables in both analyses were identified by cluster through cluster analysis. When the main effect of a group was found, multiple comparisons were conducted using Tukey's honestly significant difference (HSD) test.

Descriptive statistics and cluster analyses

Table 1 shows the means and standard deviations of and correlations among the studied variables. Cluster analysis was conducted. Examination of the tree diagram, the interpretability of each cluster, and the homogeneity of the number of participants belonging to each cluster indicated that the optimal number of clusters was four. Figure 1 shows the Z -scores of the three social achievement goals in each cluster. The first cluster (n = 25) was labeled the low cluster because the scores of all three types of social achievement goals were low. The second cluster (n = 60) was labeled the low social demonstration-approach goal cluster because the scores of the social development, social demonstration-approach, and social demonstration-avoid goals were somewhat high, clearly low, and average, respectively. The third cluster (n = 76) was labeled the high cluster because the scores of all three types of social achievement goals were high. The fourth cluster (n = 169) was labeled the average cluster because all three social achievement goals were average. Three one-way between-subjects analyses of covariance, controlling for sex, revealed that significant main effects of clusters were found for the social development goals [ F (3, 325) = 158.12, p < 0.001, η p 2 = 0.59], social demonstration-approach goals [ F (3, 325) = 134.52, p < 0.001, η p 2 = 0.55], and social demonstration-avoid goals [ F (3, 325) = 100.48, p < 0.001, η p 2 = 0.48]. Post hoc analyses indicated significant differences for both social development and social demonstration-approach goals, and significant differences were found between all pairs of clusters (all p < 0.01). For the social demonstration-avoid goals, significant differences were found between all pairs (p < 0.01), except for the low social demonstration-approach goal and average clusters (p = 0.74).

PA and NA across the clusters

The means and standard deviations of PA and NA across the four clusters, along with the results of the analyses of covariance, are presented in Table 2. A one-way analysis of covariance with PA as the dependent variable revealed a significant main effect of cluster [ F (3, 325) = 10.92, p = 0.00, η p 2 = 0.09], after controlling for sex. On average, the low cluster (M = 17.66, SE = 1.58) scored lower than the low social demonstration-approach goal cluster (M = 25.75, SE = 1.01), high cluster (M = 28.01, SE = 0.90), and average cluster (M = 26.06, SE = 0.60). Post hoc comparisons were performed using the Bonferroni correction. The difference between the low cluster and the low social demonstration-approach goal cluster, 8.10 (95% CI [3.08, 13.12]), was statistically significant (p < 0.001). The difference between the low and high clusters, 10.36 (95% CI [5.53, 15.18]), was also statistically significant (p < 0.001). The difference between the low and average clusters, 8.40 (95% CI [3.91, 12.90]), was also statistically significant (p < 0.001). However, the difference between the low social demonstration-approach goal cluster and the high cluster, 2.26 (95% CI [−1.33, 5.85]), was not statistically significant (p = 0.57). The difference between the low social demonstration-approach goal cluster and the average cluster, 0.31 (95% CI [−2.81, 3.43]), was not statistically significant (p = 1.00). The difference between the high cluster and the average cluster, 1.95 (95% CI [−0.91, 4.81]), was not statistically significant (p = 0.43).

Next, a one-way analysis of covariance with NA as the dependent variable revealed a significant main effect of cluster [ F (3, 325) = 8.57, p = 0.00, η p 2 = 0.07], after controlling for sex. On average, the low cluster (M = 17.88, SE = 1.74) scored lower than the high cluster (M = 26.57, SE = 0.98) and average cluster (M = 24.58, SE = 0.66). The low social demonstration-approach goal cluster (M = 21.39, SE = 1.11) scored lower than the high cluster (M = 26.57, SE = 0.98). Post hoc comparisons were performed using the Bonferroni correction. The difference between the low and high clusters, 8.68 (95% CI [3.38, 13.99]), was statistically significant (p = 0.00). The difference between the low and average clusters, 6.70 (95% CI [1.76, 11.64]), was also statistically significant (p = 0.002). The difference between the low social demonstration-approach goal cluster and the high clusters, 5.18 (95% CI [1.24, 9.12]), was significant (p = 0.003). However, the difference between the low cluster and the low social demonstration-approach goal cluster, 3.51 (95% CI [−2.00, 9.02]), was not statistically significant (p = 0.55). The difference between the low social demonstration-approach goal cluster and the average cluster, 3.19 (95% CI [−0.24, 6.62]), was not statistically significant (p = 0.08). The difference between the high cluster and the average cluster, 1.99 (95% CI [−1.16, 5.13]), was not statistically significant (p = 0.56).

Main findings

The first purpose of this study was to identify the combinations of the three types of social achievement goals that Japanese college students typically exhibit. Four clusters were identified. Consistent with Hypothesis 1, this study identified two clusters characterized by relatively low and high levels of all three goals. In the remaining two clusters, low social demonstration-approach goals characterized one, and average levels of the three goals, which differed from the findings of Michou et al. (2016) and Lee (2018) characterized the other. Lee (2018) identified two clusters, characterized by relatively high social development and social demonstration-approach goals, and another by relatively high social demonstration-approach and social demonstration-avoid goals. Michou et al. (2016) found a cluster characterized by a relatively high social demonstration approach goal. Although it is difficult to clarify why such differences were found in the studies conducted by Michou et al. (2016) and Lee (2018), the findings provide new insights into the characteristics and distribution of clusters among Japanese college students. Finding new clusters in a sample of college students demonstrated that the characteristics and distribution of clusters may vary from population to population.

The second objective was to examine the between-subgroup differences in PA and NA. Supporting Hypothesis 2, the results showed that the low social demonstration-approach cluster, the high cluster, and the average cluster scored higher on PA than the low cluster. From the viewpoint of high PA, the results suggest that having combinations of the three types of social achievement goals, rather than having low levels of all types of goals, was more adaptive. Interestingly, different pictures emerged regarding better adjustments in terms of NA. The low cluster showed the lowest level of NA. From the viewpoint of low NA, the results suggest that having low levels of all types of goals was more adaptive. One possible explanation for lower levels of PA and NA in the low cluster is that these college students might be less motivated to engage in social interactions with peers, and less likely to experience either affect. This hypothesis should be empirically examined in future research. The cluster that was most effectively adaptive was found to depend on whether the affectively adaptive state was defined as high PA or low NA.

Limitations

This study has some limitations. First, four clusters emerged; however, two of these were not in line with those reported by Michou et al. (2016) and Lee (2018). In future studies, it would be essential to explore the reasons behind these observed differences. Doing so would enhance our understanding of how typical patterns of social achievement goal combinations are formed within each population.

Second, the extent to which the findings of this study can be generalized remains unclear. The participants were recruited by a company and do not represent a representative sample of Japanese college students. Therefore, it is necessary to replicate the present findings using more diverse and larger sample sizes. This study provides the rationale for conducting such studies.

Third, this was a cross-sectional study, and it was impossible to examine the causality between social achievement goals, PA, and NA. At this stage, we cannot exclude the possibility that the associations of the clusters with PA and NA could be explained by a third variable. Experimental studies are necessary to examine causality, such that social achievement goals are manipulated, and to examine changes in PA and NA. The results provide a rationale for conducting such studies.

The purpose of this study was twofold. The first aim was to identify the combination of the three types of social achievement goals that Japanese college students typically have. Using a person-centered approach, four distinct clusters characterized by different configurations of social achievement goals were identified. The second purpose was to examine the between-subgroup differences in PA and NA. These clusters demonstrated differential patterns of PA and NA, indicating that affective adjustment varies depending on the combination of social achievement goals individuals endorse.

This study extends previous research by focusing on general affective states—PA and NA—rather than specific outcomes such as social anxiety or coping strategies, which have been examined in earlier studies (Lee, 2018; Michou et al., 2016). As central components of subjective well-being (Diener, 2000), PA and NA should be examined jointly to better understand college students' affective adjustment.

Importantly, the findings highlight the utility of applying a person-centered approach to the study of social achievement goals and affect. Although prior research using variable-centered approaches has demonstrated associations between individual goal types and affect (e.g., Shim et al., 2013; Zhao, 2022), the present study demonstrates that typical combinations of social achievement goals are differentially associated with PA and NA. This approach facilitates the identification of subgroups of students who may exhibit varying degrees of affective well-being, thereby offering a nuanced perspective on the role of social achievement goals in affective adjustment.

Overall, the current findings contribute to the literature on social achievement goals and well-being by clarifying how combinations of social achievement goals relate to affective adjustment among Japanese college students and by further reinforcing the value of person-centered analytical approaches in this field.